Wireless robots – magnetically guided and tracked by medical imaging and AI – can treat damage in previously inaccessible parts of the brain.
The brain vasculature (network of blood vessels) supplies the oxygen and nutrients needed by the central nervous system. Disruption – caused by stroke, aneurysms or brain tumours, for example – results in lost blood flow or excessive bleeding, often fatal.
But for treatment to be safe, it must be minimally invasive owing to the intricate structure and properties of the brain and vasculature.
Most current interventions are not dexterous enough.
Many rely on catheters to access vessels, increasingly difficult as they become much smaller, more tortuous and distant from the arterial entry point. This limits the precise delivery of drugs, stents, coils or therapies. The ‘stentbot’, delivered by catheter into a safe and easy-to-access vessel, can navigate further into distant smaller and tortuous vessels to deliver treatments, such as thrombolytic drugs. It can then be safely retrieved by the same catheter or stay inside the brain vessel where it can be repositioned for repairs for aneurysm treatments. The concept was extended to a ‘team’ of independently controlled robots to treat several sites simultaneously, tracked by medical imaging and controlled with the help of artificial intelligence (AI).
Metin Sitti, coordinator of the STENTBOT project, developer of a therapeutic robot
Robot design, locomotion, control and image-guided navigation
The European Research Council supported STENTBOT project set out to bridge the gap between the vascular locations reachable by conventional catheters, and those where treatment is needed. “The hollow cylindrical architecture of vascular stents, combining flexibility with relatively low blood flow resistance, inspired us to repurpose them into malleable wireless robots,” adds Sitti from Koc University, the project host. The resulting stentbot can operate in vessels approximately 0.8–1.5 mm in diameter, with its soft polymer structure produced by 3D mould printing, and cast with a magnetic composite, including small helical surface features to convert rotation into forward motion. “The robot autonomously adapts its shape to the vessels.
After reaching the target location, it can deliver tissue plasminogen activator drugs to treat acute ischaemic stroke or divert flow into an aneurysm site,” explains Sitti. The robot is controlled wirelessly using an external magnet mounted onto the gripper of a mechanics (seven-degrees-of-freedom) robotic arm.
While magnetic torque rotates the robot, magnetic forces and the robot’s interaction with the vessel wall enables navigation. A broader system – integrating actuation, medical imaging and AI – was developed to track and control the robot inside the body. Meanwhile, a ‘team’ of independently operating robots was also developed. In addition to magnetic actuation, these were controlled and tracked in real time using X-ray fluoroscopy and AI.
Towards a therapeutic robot platform
The robots were tested in 3D-printed models of vessels, reproducing challenging conditions including vessel diameters down to approximately 1 mm, with strong curvature, bifurcation angles up to 120 degrees and pulse flow speeds up to approximately 26 cm per second.
Additionally, mechanical and magnetic modelling revealed how magnetic forces, deformation, friction and fluid drag determine robot locomotion. The system was also evaluated in pig arteries. Demonstrating the deployment of ‘robot teams’ in complex 3D models, Sitti’s researchers explored how robot behaviour was influenced by material, stiffness and surface properties, alongside vessel diameter and blood flow.
AI-assisted tracking under challenging X-ray imaging conditions, including low contrast, noise, occlusion and surrounding anatomical structures, was also evaluated. “Our image-guided robotic platform is an important first step towards clinical robot-assisted medical interventions,” notes Sitti. Towards this aim, the team will next evaluate the system in animal models, while working closely with clinicians.
Prof. Metin Sitti is a Professor in the School of Medicine and the College of Engineering at Koç University. He is also the current President of Koç University and an Honorary Professor at the University of Stuttgart, Germany.
